EEVblog #84

[safarir]

Blowing multimeters, you make my day !

I also like the new music at the beginning and at the end


(Yes, I am monitoring your video folder  )

Sorry again for my bad english

[EEVblog]

Yeah, that was fun. Shame Doug only had the baby 400J unit, we were hoping to see a meter catch on fire. But we did get one to fully explode which was pretty cool.
Next video is a quick explaination from Doug about building a high voltage probe.

Dave.

[safarir]

Cool, I will build my own one so I could blow my old multimeter as soon as I get my new one

[Anders]

Would have been interesting too try the Smart Tweezers at 400J especially since this is really a “hand held” instrument!
Any volunteers, I mean someone holding it during the test? 

[alm]

This is probably vaguely similar to the CAT I 1000V test (4kV, 20 repetitions, 30 ohms source impedance), although the source impedance and duty cycle might be different. So the bottom line is that you shouldn't use any of the meters that exploded violently (which probably includes the early Fluke 70-series, depending on the velocity of that range switch) for mains or CAT I high-voltage (eg. flyback transformer) use. I'd still consider them safe enough for low-voltage work behind a power supply (eg. wall wart / transformer / battery / lab supply). I tend to touch that stuff with my bare hands, and use uninsulated tools with some random 5V arduino circuit, so I don't see a DMM without input protection as a an extra risk.

Disclaimer: I actually don't own any of those cheap DMM's, although I do own some older name-brand meters which don't have CAT ratings.

[mausball]

Dave-

Excellent destruction videos!

I too would have liked to see the big daddy unit connected. I'd also have liked to see the results of longer duration surges, like you can get from a wonky flyback circuit (as Doug mentioned when he was talking about the flyback diodes). The short pulses are nice, but I've actually run into things (one of which killed my original Fluke 73) that jumped north of 1kV for something like half a second. 500mS is long enough for the HV to crawl around inside and do a lot more damage, and to seriously stress protection systems.

I'd also like to see the results of such a test on some of the medium priced meters, say the $75-$150US range (old WaveTeks, Amprobe, B&K, TPI, etc). A lot of advanced hobbyists and professionals like myself keep a few of these mid-range meters around, both for their durability and for their improved performance. I know a few of mine are actually CAT I certified, but I'm curious about some others, like the B&K and TPI. Heck, even one of your nondestructive comparison reviews would be nifty for the midrange 

[Simon]

It might be a good idea to have CAT ratings explained, maybe a video blog on them ?

[alm]

Note that Fluke has got the source impedance for the CAT ratings wrong in several appnotes, I believe I complained about this in the past, but apparently they haven't fixed it (or the error was in multiple documents). They do get it right in some documents, though. One of their appnotes lists a source impedance of 30 ohm for every category, another lists the same source impedance for CAT III 1000V and CAT IV 600V.

[Kiriakos-GR]

Here they are, CAT standards ..  http://zone.ni.com/devzone/cda/tut/p/id/2827

Well after this "presentation" I am still scratching my head from confusion.

I do not believe that this was what most people had in mind as review of multimeters.

Even the article above  , states clearly  that any standard was imposed by having in mind the proper isolation, and safe behavior at the occasion of an transient .

Under the above clear specifications ,  the cheap multimeter that got the strike of the transient ,
and did not blow ,  its a Damn perfect safe multimeter .. simple as that.

The Fluke got toasted , the PCB become carbon , and  soon some one will just say that it just needs a new calibration and it will be fine ..    

Nice video , it will do the counters to jump high, but nothing useful in it , just wild action and not meaningful conclusion.

Personally I do not give a shit about Fluke , I am interested to find from the lot of manufacturers,
the ones who own a decent brand or .. a decent model of DMM .
Simple as that.

And I am even prepared to get the "Highway"  in order to keep my "emotional little world" in balance.    
    

[Simon]

so basically the CAT rating specifies the percentage of over voltage relative to the max safe voltage, I don't really see the point of it, it is far easier to just state the max voltage, my Rigol scope is CAT I 300 V, well to me CATI means nothing the important thing is that it will resist 300 V,

I would hope that the CAT ratings specify something like duration of overload  and/or other parameters otherwise it's a useless term to me.

Well we have already seen a blog before on the build quality of these meters, I mean you can go on for ever, at the end of the day it seems obvious that if you spend less than £30 on a multimeter it will not be much good, the shootout video showed just what can happen to poor quality multimeters for those who may have crazy ideas of trusting a cheap meter to 1000 V just because the manufacturer who has no come back says it can

[Zad]

I have to say I was really quite surprised about the way the Flukes handled the overload, especuially given that it is way below the rated limit. Particularly the way the sacrificial components weren't contained (no blast walls in that part of the case) and seem to have caused internal stress fractures within the PCB. Also surprising how relatively 'well' some of the Won-Hung-Lo crap did. They will never meter again, but Im not sure the Flukes would either. When I say well, I mean "would only cause a minor fire and loss of fewer than 5 fingers".

[Pradipta]

It is reasonably surprising to have an interim episode before the Multimeter Shootout. I expected to see Amprobe and Extech on test grounds as depicted by the Tweets. When will we see that big day?

[alm]

One important parameter that is omitted from the above table is the source impedance, which determines how much current the source will be able to deliver, i.e. the size of the bang if something shorts/arcs. This is the important difference between CAT I and CAT IV. Compare touching the HV electrode of a large CRT (with 30kV acceleration voltage) with touching a 33kV high-voltage transmission line. The former will hurt but probably not kill you (most of the danger is actually from the muscle reaction, not the current), the latter might actually cook or vaporize you. The former is CAT I, the latter is CAT IV. This is why CAT I tests are conducted with a 30ohm source impedance (peak short circuit current for CAT I 600V is 2500V/30ohm = ~83A), and CAT IV with a 2ohm source impedance (peak short circuit current for CAT IV 600V is 8000V/2ohm = 4000A).

I believe IEC 1010 only requires that the equipment should not be dangerous to the user (explode or catch fire), not that it keeps working.

CAT I 300V on your scope means that it's rated for up to 300V on CAT I (i.e. residential/office equipment on the secondary side of the power supply), so 110V (or 230V) mains is outside the specification. So it actually means more than just '300V max'.

[Simon]

ah yes i was also thinking of the source impedence I must have misread that table as being the item impedence (I really need an early night) but that still leaves us with the pulse duration to define.

i agree that the idea of the protection is to save the user not the instrument, if it were made of solid metal with dedicated protection chambers and you could buy replaceble protection circuitry it would be an ideal world but the price might not be ideal. It was quite obvious though that the cheaper meters are not that resistant, the one that blew apart and even the one that bulged a bit could have cause severe injury, perhaps dave could post the CAT specs of each meter and its reaction to give us an idea of what he was dealing with

[Simon]

hang on, so if the scope says CAT I 300 V does that mean that it will take 300 V at most or that it is the value to look for in that table which at CAT I will take up to 1500 V ?

[kc1980]

Cool video.  I too would like to see what would happen to the Meterman/Amprobe as I own one. 

The coil discharge reminded me of a neat coin-shrinking trick done by some crazies on youtube.   I'm not sure how many Joules/Watts are needed to shrink a coin.  Wacky...

[Simon]

well the coin changed shape, the diameter reduce and the thickness increased, i expect it was stretched by the forces at work

[alm]

hang on, so if the scope says CAT I 300 V does that mean that it will take 300 V at most or that it is the value to look for in that table which at CAT I will take up to 1500 V ?

It's designed for CAT I circuits with a continuous voltage of 300V, and will not endanger the user in case of transients (IIRC 20 pulses with a certain duration, duty cycle and source impedance) up to 1500V.

[Simon]

thats comforting, so with a 10:1 probe attached thats 15'000 volts, now I'm feeling a bit happier about probing for back EMF 

[kc1980]

well the coin changed shape, the diameter reduce and the thickness increased, i expect it was stretched by the forces at work

yes, the volume is the same so technically it's not "shrinking".  here's the best explanation that I can find, though I'm still not 100% satisfied with it:
http://capturedlightning.com/frames/shrinker.html

In the video, I completely understand why the aluminum block is repelled.  The increase/change in magnetic field from the coil induces an eddy current that generates an almost-equal but opposite magnetic field from the block.  The magnetic repulsion sends the block flying in the air.

Now, the coin shrinking thing is a bit more tricky to understand.  The assumption is that this trick works on non-magnetic metals.  I'm wondering if the coin material needs to be magnetizable.

[alm]

thats comforting, so with a 10:1 probe attached thats 15'000 volts, now I'm feeling a bit happier about probing for back EMF 

Assuming the probe is CAT I 3kV rated, and only at low frequencies. Plus it doesn't have to survive, just not blow up in your face.

[Kiriakos-GR]

hang on, so if the scope says CAT I 300 V does that mean that it will take 300 V at most or that it is the value to look for in that table which at CAT I will take up to 1500 V ?

Yes thats the idea of this table.
So any test to it, it should limited to an 1500V transient .

Generally speaking , we are unable to predict the maximum voltage of an transient , so to feel safe 100% about our DMM .  
The transient in the Mains , could happen because of a lightning absorbed by a pole , or by an 20KV transformer who had sorted out.
Its an extreme situation , totally unpredictable.

Looks logical to us , to separate the DMM's   by deferent quality standards, still no one are able to offer protection , at the next step over the Mains of 220 V, that is the 20KV !!
So the 20KV its the next enemy who could get in the 220 lines , even by an accident.

The electrical distribution network , are build by professionals, so there is special systems ( in simplified  description "Fuses" ) , that can eliminate an such scenario from happening.          
 
An Industrial called DMM are always closer in the " line of fire " ,  If we need to compare apples with apples , we should zap only DMM's called as Industrials , and the upcome will save many life's of Industrial electricians that they using them ..

Zapping DMM's for just home use , makes a nice sound and sparks , but thats all about it.

I was hopping too, the test to be limited at the max rating as limit, set by the manufacturer.
An 600V rated DMM to stay on  measuring 600V for 24 hours , its an normal test to prove it as honest.

Any way , the web its an huge space , and any one who needs or seeks information about anything, he just have to combine any information available , and make his decision at the end.

Dave did his part , and its just another deferent view of things, Biased or not , extreme or not ,
its just another view.      

[EEVblog]

I was hopping too, the test to be limited at the max rating as limit, set by the manufacturer.
An 600V rated DMM to stay on  measuring 600V for 24 hours , its an normal test to prove it as honest.

I now have a 1200V power supply adjustable in 1V steps, so can now test meters at their maximum rated voltage for however long. Look for that in all future meter reviews.

Dave.

[Pradipta]

People might find these useful...

[Simon]

thats comforting, so with a 10:1 probe attached thats 15'000 volts, now I'm feeling a bit happier about probing for back EMF 

Assuming the probe is CAT I 3kV rated, and only at low frequencies. Plus it doesn't have to survive, just not blow up in your face.

yes true the probes are rated for 600 V but I'm only measuring transients or back EMF, if the probe was damaged I would not be so bothered, it is the scope i want to stay intact.

so does CATI 300 V mean that it will withstand a continuous voltage of 300 V even though the max input is 40 V ?

[RayJones]

The modulator in our radar near Dave's workplace can deliver 2400A, 600V, 2us pulses @ 500Hz.

I figure that works out to 1440Joules (simply multiply the lot together?)

After the step up transformer, there is close to 60kV applied to the klystron tube (at a reduced current of course)

I reckon that 60kV pulse could really blow some things apart!

[alm]

yes true the probes are rated for 600 V but I'm only measuring transients or back EMF, if the probe was damaged I would not be so bothered, it is the scope i want to stay intact.

Lets hope that the probe fails open circuit, not short circuit .

so does CATI 300 V mean that it will withstand a continuous voltage of 300 V even though the max input is 40 V ?

I don't think the equipment ever has to work for IEC 1010, IEC 1010 is purely about not injuring the user. So the CAT ratings are about not exploding or catching fire. There should be a separate max non-destructive input voltage rating spec in the manual, this is the limit that it should actually survive without damage. This limit is likely to be higher than the max. vertical range, but may not be 300V.

[Kiriakos-GR]

For me, the greatest transient that came close enough to damage my isolation,
as thinking unit, was this look ..    

Its a mixture of Star Trek look, with Australian glamor    

I like teasing people ... I do not know if it works in this case  

[Neilm]

The CAT rating system is more or less as following:

CAT IV. The wiring from the power station, the voltage distribution, transformers etc up to the input of the electricity meter.
CAT III. The wiring from the meter to distribution board, large switch gear and high power outlets.
CAT II. Internal wiring up to the mains plug in the wall.
CAT I. Every thing powered through a transformer from a CAT II source.

Multi-meters should be tested to IEC61010. This calls for transient tests on the inputs of a mains voltage from a low impedance source with a transient super imposed on it. For instance a 600V CAT III meter should be tested with an 6kV transient superimposed on 600V capable of delivering at least 3000 Amps.

The IEC61010 Ed 2 is a bit vague on should the instrument still work. Most from reputable companies will work. (I would guess the resistor that was damaged on the Fluke lead to problems when the transient happened). I believe that the soon to be published Ed 3 will require the instrument to be undamaged.

I have read comments along the lines of "I've had things explode in my hand and have never had a problem". A couple of years ago I tested a 2 pole voltage probe rated 600V CAT III. It blew apart, with flash damage on the safety case we test the unit in. It also kicked and broke the clamp that held it in place.

[Kiriakos-GR]

CAT II. Internal wiring up to the mains plug in the wall.

About the mains plug in the wall , light switches , and all sort of parts ...
I have never seen  anyone from the manufacturers , to comply in anything as CAT standards.

What I am trying to say are , that even if any multimeter can pass ( or survive ) CAT testing,
the  220V installation will fall down first from an extreme over-voltage .
The weakest part in the link , as an cheap wall plug , of bulb holder , it will become toasted , even before the multimeter comes close to the maximum CAT rating as device.

Are we a bit overreacting in this CAT limits ?   

[jahonen]

Are we a bit overreacting in this CAT limits ?    

Multimeter is (could be) hand-held device, plugs and switches on the wall are not. Thus the safety requirements.

Regards,
Janne

[alm]

One difference between a lamp socket and a multimeter is that you often hold the DMM in your hands, something you rarely do with a lamp. And mains equipment can be certified (eg. UL/TUV, often required by insurance) and does have to comply to regulations (NEC in the US), although I've no idea what actually means compared to IEC 61010. Plus the fact that you are measuring something often means that something is wrong/changed.

Transients are (at least in this part of the world) a rare occurrence, but it would suck if a multimeter blew up in your face, even if it's only once. It's kinda like seat belts in a car: cars don't usually collide, but if something happens, it's nice to not have your face smash into the windshield. It's not the probability, but the severity. But the lower the energy levels and the less you use it, the lower the risk (as in probability * severity), so at some point, the risk becomes low enough that it's not worth worrying about.

[Simon]

CAT II. Internal wiring up to the mains plug in the wall.

About the mains plug in the wall , light switches , and all sort of parts ...
I have never seen  anyone from the manufacturers , to comply in anything as CAT standards.

What I am trying to say are , that even if any multimeter can pass ( or survive ) CAT testing,
the  220V installation will fall down first from an extreme over-voltage .
The weakest part in the link , as an cheap wall plug , of bulb holder , it will become toasted , even before the multimeter comes close to the maximum CAT rating as device.

Are we a bit overreacting in this CAT limits ?   

no if greek wiring standard are anything like the italian ones i experienced I'm hardly surprised 

[Kiriakos-GR]

It's not the probability, but the severity. But the lower the energy levels and the less you use it, the lower the risk (as in probability * severity), so at some point, the risk becomes low enough that it's not worth worrying about.

Ok, I agree too with this point of view ..


Now , about the "hand held" risky devices, we should pass from CAT testing all the mobile phones too,
some of them does blow at your face .. or so 

[Kiriakos-GR]

no if Greek wiring standard are anything like the Italian ones i experienced I'm hardly surprised 

Nag and Dance as much you like, me and my European brothers , we are fearless ..
We eat cables and spit sparks for breakfast .. 

[jahonen]

Now , about the "hand held" risky devices, we should pass from CAT testing all the mobile phones too,
some of them does blow at your face .. or so  

There are relevant safety standards regarding the mobile phones too. Any 230 VAC mains connected device must have an isolation which can withstand at least something like 2-3 kV between mains input (for some time, say 1 minute) and user touchable conducting part. So mobile phone charger must also withstand that. You have probably heard about exploding fake mobile phone batteries, so even with no mains at all, there are safety issues.

What is wrong with having a product safety? That is one parameter which should not be compromised, ever.

Regards,
Janne

[Kiriakos-GR]

What is wrong with having a product safety? That is one parameter which should not be compromised, ever.

Regards,
Janne

The wrong its not the device it self , its the consortium of the few who had create the high standards,
and try to dominate the market , by blaming any one who is not part of it.
If you sell safety , then everything else than yours , it must be baptized unsafe , so to make money just from that.

Or every medium scale manufacturer , he has to pay the price in the consortium, so his products to be baptized as safe too.

I do not know if you understand my point , but looks that the high safety standards ,
to be just a merchandise .

So I am equally skeptic and for the cheap PRC  made product ... and for Mister Fluke , the safety seller.

And I truly need from one True electronics engineer , to take apart one Fluke 87 and evaluate the cost of the inner parts .... I need to know where my money goes ..     

[Kiriakos-GR]

Digital cameras Flash unit as CAT tester ... not bad idea   

My Olympus FL-40 have even adjustable output to 40 steps .... But costs too much to become as DMM CAT tester ..   

[Pradipta]

Digital cameras Flash unit as CAT tester ... not bad idea  

My Olympus FL-40 have even adjustable output to 40 steps .... But costs too much to become as DMM CAT tester ..  

Men, I do not think it to be a bright idea, but we can acid test our own instrument with a car ignition coil (for all petrol engines). Connect the primary of the ignition coil to the secondary of a high power (Volt-Ampere) or high current (2 Ampere is a good compromise) rated 12 Volt mains transformer (the primary of the mains transformer being connected to the wall output). I apprehend the secondary of the ignition coil to be at greater than 1000V.
I could tell the difference between a Made in Taiwan R.O.C. multimeter and a Made in USA Fluke. No not a splendid display of fireworks I suppose but a conclusive boundary line drawn to indicate the maximum voltage tolerated by a multimeter without running out of operation (use 6V, 9V, 12V, 15V rated transformer in steps).

[Simon]

Digital cameras Flash unit as CAT tester ... not bad idea   

My Olympus FL-40 have even adjustable output to 40 steps .... But costs too much to become as DMM CAT tester ..   

the voltage will not vary the time the flash lasts will vary, photography is albout exsposure time.

I don't know about Greece but in Italy, statisticly 4.5 people a day are killed at work. now i know many of these are building site accidents (like a crane collapsing because it was used until it rusted) and the odd farmer falling into his own well and drowing but it shows the general attitude of some countries to safety. True being over saftey concious can ruin things but buying a £5 meter knowing that the CAT rating on it is probably a fake is not clever.

[Simon]

well of the three cheap meter dave tested one actually blew apart and the other bulged and emited flames (for an instant), normally you would not encounter 4 Kv but it does show just what a difference lack of protection in cheap meters can make

[Kiriakos-GR]

I don't know about Greece but in Italy, statisticly 4.5 people a day are killed at work.

We prefer to get killed in holidays and weekends with our cars , its more fun this way  

About the flash unit , what this Australian guy had made , its a flash unit in a large scale,
and nothing more ..

[saturation]

Yes, I think the bottom line in CAT ratings is safety, not survivability of the device.

One important parameter that is omitted from the above table is the source impedance, which determines how much current the source will be able to deliver, i.e. the size of the bang if something shorts/arcs. This is the important difference between CAT I and CAT IV. Compare touching the HV electrode of a large CRT (with 30kV acceleration voltage) with touching a 33kV high-voltage transmission line. The former will hurt but probably not kill you (most of the danger is actually from the muscle reaction, not the current), the latter might actually cook or vaporize you. The former is CAT I, the latter is CAT IV. This is why CAT I tests are conducted with a 30ohm source impedance (peak short circuit current for CAT I 600V is 2500V/30ohm = ~83A), and CAT IV with a 2ohm source impedance (peak short circuit current for CAT IV 600V is 8000V/2ohm = 4000A).

I believe IEC 1010 only requires that the equipment should not be dangerous to the user (explode or catch fire), not that it keeps working.

CAT I 300V on your scope means that it's rated for up to 300V on CAT I (i.e. residential/office equipment on the secondary side of the power supply), so 110V (or 230V) mains is outside the specification. So it actually means more than just '300V max'.

[saturation]

I'd like to see close ups of the damaged areas.

Also, its well known those glass fuses are inadequate for transients, how did they fair in the cheap DMMs?

What happened to any HRC and SIPA fuses, did they blow too?

Yeah, that was fun. Shame Doug only had the baby 400J unit, we were hoping to see a meter catch on fire. But we did get one to fully explode which was pretty cool.
Next video is a quick explaination from Doug about building a high voltage probe.

Dave.